Review

The Pharmacology of CD38/NADase: AnEmerging Target in Cancer and Diseases ofAging

Eduardo N. Chini,1,* Claudia C.S. Chini,1 Jair Machado Espindola Netto,1 Guilherme C. de Oliveira,1

and Wim van Schooten2

Recent reports indicate that intracellular NAD levels decline in tissues duringchronological aging, and that therapies aimed at increasing cellular NADlevels could have beneficial effects in many age-related diseases. The proteinCD38 (cluster of differentiation 38) is a multifunctional enzyme that degradesNAD and modulates cellular NAD homeostasis. At the physiological level,CD38 has been implicated in the regulation of metabolism and in thepathogenesis of multiple conditions including aging, obesity, diabetes, heartdisease, asthma, and inflammation. Interestingly, many of these functions aremediated by CD38 enzymatic activity. In addition, CD38 has also beendentified as a cell-surface marker in hematologic cancers such as multiplemyeloma, and a cytotoxic anti-CD38 antibody has been approved by the FDAfor use in this disease. Although this is a remarkable development, killingCD38-positive tumor cells with cytotoxic anti-CD38 antibodies is only one ofthe potential pharmacological uses of targeting CD38. The present reviewdiscusses the biology of the CD38 enzyme and the current state of develop-ment of pharmacological tools aimed at CD38, and explores how these agentsmay represent a novel approach for treating human conditions includingcancer, metabolic disease, and diseases of aging.

Emerging Roles of NAD Metabolism in Human Disease: The Promise of‘NAD Boosting’ TherapiesNAD is a cofactor in electron transfer during oxidation/reduction reactions, and also plays acrucial role in cell signaling, regulating several pathways from intracellular calcium transients tothe epigenetic status of chromatin [1–6]. Thus, NAD is a molecule that provides a link betweensignaling and metabolism. Importantly, a decline in cellular NAD levels has emerged as apotential key player in the pathogenesis of several diseases including age-related conditions(Table 1) [1–32].

The age-related decline of NAD is of particular interest [1–6]. As the world population ages, theincidence of age-related comorbidities continues to increase. It appears that age-related organdysfunction and age-related diseases are, at least in part, governed by impairment in tissueenergy metabolism [1–14]. Understanding the mechanisms that drive these metabolic abnor-malities in aging may lead to the development of new therapies for the elderly population.Unfortunately, to date there are no therapies directly aimed at any of the fundamental biologicalmechanisms of aging [33]. In animal models, age-related metabolic diseases and decline in

HighlightsChanges in NAD metabolism play animportant role in the aging process andin the pathogenesis of severaldiseases.

‘NAD boosting’ therapy can promoteincreases in longevity and healthspanin animal models of aging, acceleratedaging, and age-related disease.

CD38 is one of the main NAD-degrad-ing enzymes in mammalian tissues,and plays a key role in age-relatedNAD decline.

CD38 is a druggable target in the ther-apy of human cancers.

Inhibition of CD38 with small mole-cules or monoclonal antibodies candecrease NADase activity and boostcellular NAD levels.

1Signal Transduction and MolecularNutrition Laboratory, Kogod AgingCenter, Department of Anesthesiologyand Perioperative Medicine, MayoClinic College of Medicine, Rochester,MN 55905, USA2Teneobio Inc., 1490 O’Brien Drive,Menlo Park, CA 94025, USA

*Correspondence:chini.eduardo@mayo.edu (E.N. Chini).

424 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 https://doi.org/10.1016/j.tips.2018.02.001

© 2018 Elsevier Ltd. All rights reserved.

GlossaryAdenosine diphosphoribose(ADPR) and cyclic ADPR(cADPR): second messengersinvolved in calcium signaling.ADP-ribosyl cyclase: an enzymethat catalyzes the conversion of NADinto its cyclic analog cADPR andnicotinamide.NADase: an enzymatic activity thatleads to the degradation of NADirrespective of the products.NAD glycohydrolase: enzymecatalyzing the cleavage of theglycosidic bond between thenicotinamide and ribose moieties ofNAD. This leads to the degradationof NAD and formation of theproducts ADPR and nicotinamide.NAD precursors: molecules thatcan be used by cells and tissues tomake NAD. They includenicotinamide and vitamin B3derivatives such as nicotinamideriboside (NR) and nicotinamidemononucleotide (NMN).Sirtuins: NAD-dependentdeacetylases involved in the controlof metabolism, healthspan, andaging. In mammals there are sevensirtuins (SIRT1–7).

healthspan can be mitigated by therapies that restore tissue levels of NAD [1–32], suggesting arole for ‘NAD boosting’ therapies [1–31].

Experimental NAD boosting or NAD replacement can be accomplished via administration ofNAD precursors (see Glossary) such nicotinamide mononucleotide (NMN), nicotinamideriboside (NR), and derivatives of vitamin B3, or via inhibition of NAD-degrading enzymes[1–31]. Therefore, it is important to determine the mechanisms that regulate the homeostasisof this nucleotide in vivo. The reactions that consume NAD are mediated by enzymes involved inthe non-oxidative roles of NAD [1–4]. Of these enzymes, CD38, PARPs, and SARM1 appear tosignificantly contribute to cellular NAD degradation [1–4,6,19,28,34–39]. Until recently theprevailing hypothesis to explain the age-related NAD decline was that activation of DNA repairenzymes such as PARPs during the aging process would consume and deplete cellular NAD[1–4]. However, we demonstrated that levels and activity of CD38 increase during aging, andthat this enzyme is, at least in part, responsible for the age-related NAD decline [6]. Forexample, genetic ablation of CD38 protects against age-related NAD decline and mitochondrialdysfunction [6]. Given the emerging role of NAD homeostasis in aging and age-relateddiseases, our discovery of the role of CD38 in the control of cellular NAD levels opens thepossibility that CD38 inhibition may be used as a therapeutic approach to increase cellular NADlevels (Figure 1, Key Figure) [1–4,6]. We review here the biological, biochemical, and pharma-cological aspects of CD38, with emphasis on its role in NAD homeostasis.

CD38: From Cell-Surface Marker to Multifunctional Enzyme Involved in NADMetabolismCD38 was first identified by the Herculean task of E.L. Reinherz and S.F. Schlossman tocharacterize and identify cell-surface markers with monoclonal antibodies as part of theirpioneering search for molecules for immunophenotyping of T cells [38,39]. However, beyonda role as structural marker, CD38 is involved in several physiological and pathologicalconditions [6,19,34–40]. In particular, CD38 is the main NAD-degrading enzyme in severalmammalian tissues [6,19,34–40]. Some pharmacological approaches to inhibit CD38 takeadvantage of its role as a cell-surface marker when the intention is to destroy or modulateCD38-positive cancer and/or immune cells, as in multiple myeloma (MM) and other cancers(Figure 2) [41–50]. By contrast, modulating CD38 enzymatic activity may serve as a ‘NADboosting’ therapy for aging and metabolic diseases (Figure 1) [1–4,6,19,36,39]. We discuss

Table 1. Conditions Where Cellular NAD Decline or the Beneficial Effects of ‘NAD Boosting’ Therapy HaveBeen Described

Conditions Refs

Aging, longevity, healthspan, and progeroid syndromes [5–14,16]

Cataract, genetic macular degeneration [14,15]

Hearing loss [16,17]

Obesity and diabetes [8,18–20]

Non-alcoholic and alcoholic fat liver disease [19,21–23]

Kidney and heart disease [24,27]

Neurodegeneration-related disease and stroke [28,29]

Muscular dystrophy and muscular mitochondrial disease [30,31]

Fetal malformation (Vactrel syndrome) [32]

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 425

Key Figure

‘NAD-Boosting’ Therapy via CD38 Inhibition in Age-Related Diseases

(A)

(B)

cADPR

CD38

Aging‘Infalmmaging’

CD38/NADase NAD

Aging phenotype

Age-related diseases

Metabolic dysfunc on

Healthspanand longevityNAD

CD38i(smCD38i and an -CD38 mAb)

NAM

ADPR

Cyclase

Hydrolase

N-ribosylbond

NAMβ-NAD

(See figure legend on the bottom of the next page.)

Figure 1. (A) CD38 cleaves NAD into ADPR/cADPR and NAM. The main site of reaction in the NAD molecule is the terminal ribose, specifically the covalent bondbetween N1 in nicotinamide and the anomeric carbon of the ribose. CD38 cleaves the N-glycosyl bond, and the main products of the reaction are ADPR andnicotinamide, with a very small proportion of NAD being converted to cADPR. Abbreviations: ADPR, adenosine diphosphoribose; cADPR cyclic adenosine

426 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4

below key biological aspects of CD38 and the potential pharmacological tools available tostudy and target this molecule, including anti-CD38 antibodies and small-molecule inhibitors.

The Biology of CD38CD38 is a multifunctional protein involved on cellular and tissue NAD homeostasis, and also inthe generation of the second messengers adenosine diphosphoribose (ADPR) and cyclic-ADPR (cADPR) that are involved in intracellular calcium signaling [38–40,51]. Functional andstructural data indicate that the promoter region of the human CD38 gene, located onchromosome 4, is regulated by several factors including NF-kB, RXR, LXR, and STAT[38,39,52]. Importantly, CD38 expression is induced by inflammatory cytokines, endotoxins,interferon, and some nuclear receptors (Figure 3) [38,39,52]. Therefore, TNF-a and lipopoly-saccharide (LPS) treatment results in increased expression of CD38/NADase activity and asubsequent dramatic decrease in cellular NAD levels [38,39,52]. Understanding the mecha-nisms regulating CD38 gene expression provides clues to the mechanisms governing theincreased expression of CD38. In particular, the proinflammatory state observed during theaging process may provide a link between cytokine-induced CD38 expression, age-relatedtissue NAD decline, and the ‘inflammaging’ theory [1,53].

A unique characteristic of the CD38 enzyme is its cellular localization. The great majority ofCD38 has a type II membrane orientation, with the catalytic site facing the outside of the cell[38,39]. Thus, >90% of CD38 enzyme functions as an ecto-NADase [54]. Because the majorityof substrates for CD38 NADase are expected to be intracellular, the ecto-enzymatic activity ofthis enzyme represents an intriguing ‘topological paradox’ (Figure 3B). Until recently it had beendifficult to explain this paradox. However, we and others demonstrated that CD38 degradesnot only NAD but also circulating NAD precursors such as NMN and NR before they can beincorporated into cells for NAD biosynthetic pathways [6,55].

CD38 has also been observed in intracellular membranes, such as in the nuclear membrane,mitochondria, and endoplasmic reticulum [38,39]. In addition, a small fraction of CD38 is alsoexpressed as a type III plasma membrane protein with the catalytic site facing the inside of thecell (Figure 3B) [56]. Finally, soluble intra- and extracellular forms of CD38 have also beendescribed (Figure 3) [38,39,57]. The relative roles of the different ‘anatomical’ locations ofCD38 in cellular and tissue NAD homeostasis is one of the outstanding key questions in thefield.

Catalytic Activity and Physiological Roles of CD38As discussed above, CD38 is a multifunctional enzyme that uses b-NAD and other b-NADderivatives, such as b-NADP and b-NMN, as substrates. Interestingly, CD38 does not accepteither a-NAD or NADH as substrates. The primary catalytic reaction of CD38 involves thecleavage of a high energy b-glycoside bond between nicotinamide and the ribose moiety.During catalysis, the removal of the nicotinamide moiety from b-NAD is coupled with theformation of non-covalent enzyme (E)–substrate intermediates such as E-ADPR or the less

diphosphoribose; NAM, nicotinamide. (B) A proinflammatory phenotype is observed during the aging process (inflammaging hypothesis). Inflammatory cytokines andendotoxins are potent inducers of CD38 expression, leading to an increase in tissue NADase activity and a subsequent decline in cellular NAD levels that may play a rolein the development of the aging phenotype, decrease resilience, metabolic dysfunction, and the appearance of some age-related diseases. Inhibition of CD38 via small-molecule CD38 inhibitors (smCD38i) or non-cytotoxic monoclonal antibodies that inhibit CD38 activity (CD38i mAb) may prevent cellular NAD decline and promotehealthspan and successful aging. Symbols: #, decrease; ", increase.

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 427

abundant E-cADPR. These intermediates are stabilized through H-bonds between their ribosyl20,30-OH groups and the catalytic residue Glu226, a residue required for the NADase andcyclase activity of the enzyme [38–40,58–60]. Structural and mutagenesis experimentsrevealed other important residues in the catalytic site, including Glu146, Trp189, Trp125,Ser126, Thr221, and Arg127 [38–40,59,60]. The intermediates of the reaction are releasedfrom the catalytic site to generate ADPR or cADPR. In general, the majority of the CD38NADase catalytic activity will generate nicotinamide, and a nearly stoichiometric amount ofADPR, with residual amounts of NAD being converted to cADPR and nicotinamide [34,35,38–40,58–60]. These two catalytic activities of the enzyme are classified as NAD glycohydrolaseand ADPR-ribosyl cyclase [34,35,38–40,51,58–60]. The amino acid sequence of CD38 hasconsiderable homology with a pure ADP-ribosyl cyclase from Aplysia [38–40,60]. At least tworesidues in CD38 contribute to its increased glycohydrolase versus cyclase activity, namelyThr221 and Glu146 [58–60]. In fact, mutagenesis of these residues abolishes CD38 NADaseactivity and increases its cyclase activity [60].

Both ADPR and cADPR have second messenger signaling roles [38–40,51,58,60]. The roles ofCD38 as a cyclase and of NAD-derived calcium messengers in physiology and pathology havebeen extensively reviewed [38–40,51,58,60]. Importantly, it should be highlighted that whenCD38 catalyzes the degradation of NAD to ADPR, or the cyclization of NAD to cADPR, there isNAD consumption and nicotinamide generation, and CD38 therefore plays a crucial role intissue NAD and nicotinamide homeostasis [6,34,35,38–40].

Importantly, owing to its ecto-enzymatic activity, CD38 not only regulates intracellular andextracellular NAD homeostasis but also modulates the availability of extracellular NAD and itsmetabolites [6,57]. For example, CD38 ecto-NADase activity plays a role in the extracellularadenosine pathway [61]. In fact, it was reported that extracellular adenosine can be generatedvia both NAD and ATP degradation via a cascade of events that include CD38 ecto-NADaseand CD39 ecto-ATPase activity. In turn, adenosine can bind to specific purinergic P1 receptors,

smCD38i or an -CD38 an body

Cytotoxic effects

Immune cell an -tumor response

CD38+ immune cell CD38+ cancer cell

NADase inhibi on

Tumor celldeath

CD38

CD38

CD38

Nucleus

NAD

Figure 2. CD38 Targeted Therapy in Cancer. CD38 has at least two potential roles in cancer therapy. First, cytotoxicanti-CD38 antibodies can promote the killing of CD38-positive cancer cells via direct and indirect effects. By contrast,CD38 expression in immune cells, cancer cells, and other cells in the tumor microenvironment may cause a decrease intissue NAD levels that has a negative effect on the immune response against tumor cells. Thus, small-molecule CD38inhibitors (smCD38i) and monoclonal antibodies that inhibit CD38 activity (CD38imAb) may promote an increase in tissueNAD levels and a positive antitumor immune response.

428 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4

Endotoxins and cytokines(TNF-α, IL-6)

Interferon-γ

LXR ligands(25-HC)

Plasmamembrane

Extracellular

Extracellular

Intracellular

Intracellular

Ecto-cataly c site

Endo-cataly c site

An body

NAD

NAD/NMN

Type IIICD38

Type II

CD38

iCD38

esCD38

CD38

CD38

CD38mRNA

CD38 gene

ADPR + NAMNucleus

NAD

STAT

LXRNF-κB

NAD/NMN

ADPR/RMP + NAM

Ecto-NAD/NMNasecataly c domain

Endo-NAD/NMNasecataly c domain

‘Inflammaging’

(+)

(+)

(+)

(A)

(B)

Figure 3. The Biology of CD38 and the Topological Paradox. (A) Regulation of CD38 gene expression byinflammatory agents and the link between ‘inflammaging’, increased CD38 expression, and cellular NAD decline.CD38 expression is activated by lipopolysaccharide (LPS), cytokines, interferon, and LXR ligands such as 25-hydro-xycholesterol (25-HC). In particular, cytokines, interferon, and endotoxins have been proposed to play a role on the sterileinflammatory process observed during aging. In (B) the CD38 topological paradox dictates that, although most NAD isintracellular, only a minority of CD38 has its catalytic site facing the inside of the cells. Thus, the most prevalent form ofCD38 (type II), that has its catalytic site facing the outside, degrades extracellular NAD and NAD precursors, such as NMN,limiting their availability for NAD synthesis in intracellular pools. A secretory soluble form of CD38 with extracellular NADaseactivity has also been described (esCD38). By contrast, the type III transmembrane form of CD38 has its catalytic sitefacing the inside, hence having access to the larger intracellular NAD pools. Minor expression of CD38 has also beendescribed in intracellular membranes such as the nuclear membrane and mitochondrial membrane. The expression ofintracellular CD38 may be limited, to prevent cellular NAD degradation and cellular metabolic collapse. Abbreviations:NAM, nicotinamide; RMP, ribose monophosphate.

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 429

eliciting signals in multiple cells including immunomodulatory effects in the tumor microenvi-ronment [61].

Very importantly, CD38 has a crucial role in the regulation of NAD-dependent processes inseveral cellular compartments. For instance, CD38 regulates the activity of the nuclear andmitochondrial sirtuins [6,34,35]. Sirtuins are NAD-dependent deacetylases implicated in theregulation of metabolism and several aspects of ageing, healthspan, and longevity [2,3]. Inparticular, we observed that CD38 degrades NAD and decreases the accessibility of NAD tothese enzymes [6,34,35]. It is also possible that generation of nicotinamide by CD38 mayregulate sirtuin activity because nicotinamide is an endogenous inhibitor of the sirtuins [62,63].Thus, the potential role of CD38 as a sirtuin regulator is of major pharmacological relevance.

Pharmacological Approaches to Targeting/Modulating CD38The identification of CD38 as a key enzyme involved in NAD metabolism, cell signaling, andcancer suggests its potential as a target in multiple conditions. Based on the therapeutic goal,CD38 can be targeted using different pharmacological approaches such as cytotoxic anti-bodies, enzyme-blocking antibodies, and reversible or irreversible small-molecule inhibitors.

Anti-CD38 Antibodies: A Success Story in Multiple Myeloma and Future Perspectives inNAD Replacement TherapyCD38 is highly and uniformly expressed at the cell surface of MM cells, and at lower levels innormal lymphoid and myeloid cells, as well as in some tissues of non-hematopoietic origin[38,41–49]. This has spurred the development of several anti-CD38 antibodies for MM [41–49].In fact, monoclonal antibodies (mAbs) against CD38 are highly efficacious in the treatment ofMM [41–49]. Currently, four mAbs are in clinical trials for the treatment of CD38+ malignancies[41–49]. The most advanced is daratumumab (Janssen Biotech) which received FDA approvalfor MM in 2015 [41,42]. Three other CD38-binding antibodies are in clinical trials: isatuximab(Sanofi-Aventis), MOR202 (Morphosys), and TAK079 (Takeda) [42,48,49,64] (Table 2). Direct invitro comparisons of these mAbs showed comparable antibody-dependent cell-mediatedtoxicity (ADCC) and binding affinities, but remarkable differences in their ability to induce directapoptosis, to induce complement-mediated cytotoxicity (CDC), to inhibit enzymatic activities,and to induce antibody-dependent cell-mediated phagocytosis (ADCP) (Figure 2 and Table 1)[41,42]. In addition, it has been demonstrated that these anti-CD38 mAbs also have a potentialin vivo immunomodulatory effects on the tumor microenvironment including enhancing effectorT cell function and inhibiting suppressive regulatory T cell (Treg) activity (42). In light of the factthat these three anti-CD38 antibodies show similar safety and efficacy profiles, it is hypothe-sized that ADCC is the main mechanism of action of these antibodies in MM [41,42]. However,as discussed above, CD38 is a multifunctional membrane enzyme and regulates a variety ofNAD-dependent cellular processes. Although these antibodies were selected on the basis ofcytolysis, it is possible that some of their therapeutic effects may be mediated by inhibition of theNADase activity and subsequent NAD boosting effects. In particular, the immunomodulatoryeffects of anti-CD38 antibodies during cancer therapy may be at least in part related to thedecrease in CD38 NADase activity. Very recently, Chatterjee et al. showed that the CD38–NADase–NAD+ axis plays an important role in the immune response of T cells in a preclinicalmodel of melanoma [50]. These studies indicate that high levels of NAD+, negatively regulatedby CD38, preserve T cell function against tumor cells, raising the possibility that inhibition ofCD38 may work synergistically with blockade of PD-1/PD-L1 pathway in immune therapy forcancer. These findings suggest that combined therapy may lead to superior tumor responses.To date, isatuximab is the only clinically relevant anti-CD38 antibody shown to inhibit thecatalytic activity of the enzyme (Table 2). Anti-CD38 antibodies that specifically inhibit CD38

430 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4

NADase activity without cytotoxic effects may become an important tool for boosting NAD,immune modulation, and for use in age-related diseases. These new antibodies may pave theway for the development of highly specific CD38 NADase inhibitors aimed at ‘NAD boosting’therapy in the near future. However, to date there is no in vivo evidence that therapeutic anti-CD38 mAbs exert their effects via inhibition of CD38 NADase activity. In fact, it is not knownwhether the antitumor effects of inhibiting CD38 are mediated partly by inhibition of its CD38NADase activity.

Small-Molecule CD38 InhibitorsTo date over 200 compounds are listed as CD38 inhibitors in the literature [40,54,58–60,65–78]. Regarding chemical structures, CD38 inhibitors can be classified as NAD-analogs,flavonoids, and heterocyclic compounds [40,54,60,65–78]. Based on their mechanism ofaction, they can be pooled into two groups: covalent and non-covalent inhibitors (Table 2).Covalent inhibitors form a bond with Glu226 in the active site. Even though an excess ofnicotinamide can recover enzymatic activity, under physiological conditions the rate of disso-ciation is expected to be very slow [3]. By contrast, non-covalent inhibitors bind to amino acidresides in the active site of the enzyme through weaker interactions including hydrogen andhydrophobic bonds [1–4].

Table 2. Pharmacological Tools for Targeting CD38

Compound name Mechanism of action IC50 or Ki In vivo/ex vivo data Refs

Covalent Inhibitors

NAD Analogs Ara-F-NAD Competitive inhibition of NADase activity 169 nM No [66,69]

Ara-F-NMN Competitive inhibition of NADase activity 69 nM No [69]

Ara-F-NMNPhosphoester/C48

Competitive inhibition of NADase activity NAa Yes [70]

Non-covalent inhibitors

Carba-NAD Competitive inhibition of NADase activity 100 mM No [67,68]

Pseudocarba-NAD Competitive inhibition of NADase activity 186 mM No [67]

Flavonoids Apigenin Competitive inhibition of NADase activity 15 mM Yes [59,73]

Luteolinidin Competitive inhibition of NADase activity 6.0 mM11.4 mM

Yes [71,73]

Kuromanin Competitive inhibition of NADase activity 6.3 mM Yes [59,71,77]

Rhein/K-rhein Noncompetitive inhibition of NADase activity 1.24 mM0.84 mM

Yes [59,75,76]

4-Amino-quinolines 78c Competitive inhibition of NADase activity 7.3 nM Yes [78]

1ah Competitive inhibition of NADase activity 510 nM Yes [79]

1ai Competitive inhibition of NADase activity 150 nM Yes [79]

Antibodies (IgG mAb)b Isatuximab Allosteric inhibition of NADase activity NA Yes [42,49]

Daratumumab Cytotoxic effect/clearance of CD38+ cells NA Yes [41–43]

TAK-079 Cytotoxic effect/clearance of CD38+ cells NA Yes [64]

MOR-202 Cytotoxic effect/clearance of CD38+ cells NA Yes [42]

aNA, data not available.bMonoclonal antibodies have direct and indirect cytotoxic effects. Direct in vitro comparisons of these mAbs showed comparable antibody-dependent cell-mediatedtoxicity (ADCC) and binding affinities, but remarkable differences in their ability to induce direct apoptosis, complement-mediated cytotoxicity (CDC), or antibody-dependent cell-mediated phagocytosis (ADCP). Inhibition of CD38 enzymatic activity has only been reported for isatuximab.

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 431

Mechanistic/Structure-Based NAD Analogs as CD38 InhibitorsThis class of CD38 inhibitors was developed via modification of the nicotinamide ribose in themolecule of NAD and NMN [58,59,66,67]. These inhibitors were designed based on knowledgeof the mechanisms of catalysis and the crystal structure of CD38, and are considered asmechanistic/structure-based inhibitors. These molecules include carba-NAD and arabinose(Ara)-NAD analogs [54,59,66,67]. Carba-NAD and pseudocarba-NAD are non-covalent inhib-itors of CD38, and have low affinity for the enzyme (Table 2). By contrast, Ara derivatives arecovalent inhibitors of CD38, and their potency is further increased when the 2-hydroxyl group isreplaced by a fluoride (Table 2 and Figure 4). Figure 4 shows the modifications that led to theseanalogs. Likewise, modification of the nicotinamide ribose in NMN and NR led to even morepotent covalent and non-covalent inhibitors of CD38 (Figure 4 and Table 2).

The therapeutic potential of these NAD analog inhibitors may be limited due to the inhibitoryeffect they may have in several other NAD-dependent enzymes, and this raises concerns abouttheir specificity. Thus, these ‘non-specific’ effects may limit their in vivo applicability. However, itis possible that in the future some covalent inhibitors may serve as CD38 modifiers in therapiesaimed at ex vivo optimization of immune cell therapy for cancer [50]. For example, based onrecent published data, one could envisage that treatment of ex vivo hybrid Th1/17 (type 1/17 Thelper) cells with these inhibitors could increase their tumor-suppressive effects [50].

FlavonoidsSeveral natural compounds have been reported to inhibit the catalytic activity of CD38[40,71–76]. These include flavonoids such as apigenin, quercetin, and leteolinidin (Figure 4and Table 2). Beneficial effects of flavonoid CD38 inhibitors were reported in animal models ofobesity, heart ischemia, kidney injury, viral infection, and cancer [71–77]. Mechanistically,inhibition of CD38 in vivo via apigenin leads to an increase in cellular NAD levels and activationof NAD-dependent enzymes (e.g., sirtuins) [72]. Most flavonoids are competitive antagonists ofCD38, and 4,5-dihydroxyanthraquinone-2-carboxylic acid (Rhein) is the only non-competitiveinhibitor reported to date (Table 2) [75,76]. Unfortunately, flavonoids are far from being specificinhibitors of CD38, showing several ‘off-target’ effects, and also have a poor oral pharmacoki-netic profile [1,2,9–13]. By contrast, some of these compounds were described as safe forhuman use, supporting the druggability of CD38.

CD38 Inhibitors Derived from 4-Amino-QuinolineThe availability of data from crystallography of CD38, and a better understanding of itsmechanism of catalysis, have paved the road for the design of new high-affinity and morespecific CD38 inhibitors [78,79]. These small-molecule candidates are non-covalent reversibleinhibitors of the enzyme. Two studies using high-throughput screening discovered activecompounds with an IC50 in the nanomolar range [78,79]. The compounds are heterocyclicderivatives of 4-amino-quinoline [78,79]. The three lead compounds, 78c (IC50 7.3 nM), 1ai(IC50 46 nM), and Iah (IC50 115 nM) increased NAD levels in muscle and liver of diet-inducedobese mice [78,79]. They have an improved pharmacokinetic profile for oral administration, andacceptable specificity, chemical, and biological stability, compared to NAD analogs andflavonoids. Mechanistically, they may interact with other amino acid residues in the catalyticsite, rather than Glu226. For 78c, the most important interaction in the active site appears to bea hydrophobic interaction between its C4 group and Trp176 of the enzyme, and a hydrogenbond between the 2-ketoresidue and the side-chain amide of Asp183 [78]. For the otherderivatives, hydrogen bonds between the carboxylate side chain of both Glu146 and Asp155and face-to-face p–p stacking with Trp189 appear to be more important [78,79]. As previously

432 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4

OO

O

O

O

O O

O

O

O

O

O O

O

OO

O

O O

O

O

O

O

O O

O

OO

O

OHHO

OH

OH OH

OH

OH OH

OH

OHHOHO

HO

HO HO

HOHO

O O+

O

O

O

O

O

OHOH

OH

HO

HO

HO

HO

HO

PP

P

P

P

P

N N

NN

N

N

NN

NH2

NH2 NH2NH2

CONH2

CONH2

CONH2 CONH2CONH2

OHOH

OHOH

OH OH

N

N

N N

N N

NN

P P

P

HNHN

NN

N

NH2

FFF

FS

CH3

CH3

H3C

CovalentCD38 inhibitors

Non-covalentCD38 inhibitors

Ara-analogs Carba-analogs

Flavonoids

4-Quinoline

Ara-NAD

Ara-F-NAD

Apigenin

Carba-NAD Pseudocarba-NAD

Luteolinidin

Ara-F-NMN 78c 1ah

OH HO

N+

N+

N+N+N+

F

F

Figure 4. The Chemical Structure of the Different Classes of Small-Molecule CD38 Inhibitors (smCD38i). The overall mechanism of action is inhibition ofNAD breakdown. To date no inhibitors selectively block the glycohydrolase versus cyclase activities of CD38. NAD analogs can be classified as covalent and non-covalent inhibitors. Ara-NAD analogs are covalent competitive inhibitors. Carba-analogs are non-covalent competitive inhibitors. Most flavonoids and small-moleculeinhibitors of CD38 are non-covalent competitive inhibitors. Rhein is the only noncompetitive inhibitor in the flavonoid class.

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 433

discussed, mutagenesis studies revealed that these residues are important for the NADaseactivity of human CD38, and changes in these residues led to the abolition of the glycohy-drolase activity and an increase in cyclase activity. Blockade of the interaction of endogenoussubstrates with these residues thus appears to be the molecular mechanism of inhibition bythese drugs [78,79]. It is possible that these inhibitors may also have selective inhibitory effectson CD38 NADase over its cyclase activity. Thus, these compounds may become powerful toolsto understand the importance of the different catalytic activities of CD38 in a multiplicity ofconditions including asthma, cancer, metabolic syndrome, and aging. Because these drugsincrease NAD levels in many tissues in vivo, and our studies show that CD38 is the mainNADase regulating NAD levels during aging, these compounds can also improve and advance‘NAD boosting’ therapy for aging-related diseases.

Concluding RemarksThe development of non-cytotoxic anti-CD38 antibodies and small molecules targeting theNADase enzymatic activity of CD38 represent an important advance in therapies for cancersand age-related diseases, but many important questions remain about CD38 biology andpotential therapeutic uses (see Outstanding Questions). One of these questions concerns theCD38 ‘topological paradox’ and its role in intercellular NAD homeostasis. Of particular interest,is there a specific role for CD38 subcellular localization on the dynamics of intracellular NADcompartmentalization? For example, does CD38 preferentially regulate NAD levels in specificlocations such as the mitochondria or nuclei? This is a very important question that couldpotentially be approached experimentally by using CD38 constructs targeted to differentsubcellular distributions in conjunction with the newly developed NAD biosensors [80,81].Another important question in the biology of NAD and CD38 is related to the potential role of theCD38 paralog BST1 in organismal NAD homeostasis. Finally, from the pharmacological point ofview it is necessary to determine the therapeutic role of CD38/NADase inhibitors and other NAD‘boosting’ therapies for diseases of aging and metabolic syndrome. In this regard, a crucialquestion concerns the potential unwanted side effects of CD38 inhibitors? Given the role ofCD38 in the immune system and behavior, these may include impaired immune response toinfections and neurological abnormalities [40,50–52,82,83]. These side effects may differbetween different types of CD38 inhibitors because they have different mechanisms of inhibi-tion and effects on the cyclase versus NADase activities of CD38, as well as different tissuedistributions and pharmacokinetic profiles. However, the low blood–brain barrier penetrabilityof antibodies and some small-molecule inhibitors may mitigate the theoretical neurological sideeffects of CD38 inhibitors. Future preclinical and clinical studies will bring light to the potentialtransformative role of CD38 inhibitors in human disease.

AcknowledgmentsWork in laboratory of E.N.C. is supported in part by grants from the Ted Nash Long Life Foundation, the Glenn Foundation

for Medical Research via the Paul F. Glenn Laboratories for the Biology of Aging at the Mayo Clinic, and National Institutes

of Health (NIH) grants from the National Institute of Aging (NIA, grant AG-26094) and the National Cancer Institute (NCI) via

the Mayo Clinic pancreatic cancer SPORE program (CA 102701-14P2). E.N.C. has a patent on CD38 inhibitors and is a

consultant for Teneobio. W.v.S. is the Chief Scientific Officer of Teneobio, a company interested in the development of

therapeutic antibodies.

Supplemental InformationSupplemental information associated with this article can be found, in the online version, at https://doi.org/10.1016/j.tips.

2018.02.001.

Outstanding QuestionsIs CD38 ecto-NADase activity the maincontributor to NAD homeostasis invivo?

How does CD38, a predominantlyecto-NADase, regulate the homeosta-sis of NAD inside the cell?

How do the different orientations andsubcellular distribution of CD38 con-tribute to the compartmentalization ofNAD homeostasis?

Do CD38-positive cells regulate theavailability of NAD precursors toCD38-negative cells in vivo?

Does BST1, a protein with homologyto CD38, also play a role in organismalNAD homeostasis?

Will NAD boosting therapies be aneffective way to treat or prevent age-related diseases and metabolicsyndrome?

Would the combination of CD38 inhib-itors with NAD precursors present anadvantage in NAD boosting therapy forage-related diseases?

What will be the unwanted side effectsof inhibiting CD38 NADase activity?

434 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4

References1. Chini, C.C.S. et al. (2017) NAD and the aging process: role in life,

death and everything in between. Mol. Cell. Endocrinol. 455, 62–74

2. Verdin, E. (2015) NAD+ in aging, metabolism, and neurodegen-eration. Science 350, 1208–1213

3. Imai, S. and Guarente, L. (2014) NAD+ and sirtuins in aging anddisease. Trends Cell Biol. 24, 464–471

4. Schultz, M.D. and Sinclair, D.A. (2016) Why NAD+ declines duringaging: it’s destroyed. Cell Metab. 23, 965–966

5. Gomes, A.P. et al. (2013) Declining NAD+ induces a pseudohy-poxic state disrupting nuclear–mitochondrial communication dur-ing aging. Cell 155, 1624–1638

6. Camacho-Pereira, J. et al. (2016) CD38 dictates age-related NADdecline and mitochondrial dysfunction through an SIRT3-depen-dent mechanism. Cell Metab. 23, 1127–1139

7. Mouchiroud, L. et al. (2013) The NAD+/sirtuin pathway modulateslongevity through activation of mitochondrial UPR and FOXOsignaling. Cell 154, 430–441

8. Yoshino, J. (2011) Nicotinamide mononucleotide, a key NAD+

intermediate, treats the pathophysiology of diet- and age-induceddiabetes in mice. Cell Metab. 14, 528–536

9. de Picciotto, N.E. et al. (2016) Nicotinamide mononucleotidesupplementation reverses vascular dysfunction and oxidativestress with aging in mice. Aging Cell 15, 522–530

10. Zhang, H. et al. (2016) NAD+ repletion improves mitochondrialand stem cell function and enhances life span in mice. Science352, 1436–1443

11. North, B.J. et al. (2014) SIRT2 induces the checkpoint kinaseBubR1 to increase lifespan. EMBO J. 33, 1438–1453

12. Mills, K.F. et al. (2016) Long-term administration of nicotinamidemononucleotide mitigates age-associated physiological declinein mice. Cell Metab. 24, 795–806

13. Scheibye-Knudsen, M. et al. (2014) A high-fat diet and NAD+

activate Sirt1 to rescue premature aging in Cockayne syndrome.Cell Metab. 20, 840–855

14. Williams, P.A. et al. (2017) Vitamin B3 modulates mitochondrialvulnerability and prevents glaucoma in aged mice. Science 355,756–760

15. Koenekoop, R.K. et al. (2012) Mutations in NMNAT1 cause Lebercongenital amaurosis and identify a new disease pathway forretinal degeneration. Nat. Genet. 44, 1035–1039

16. Someya, S. et al. (2010) Sirt3 mediates reduction of oxidativedamage and prevention of age-related hearing loss under caloricrestriction. Cell 143, 802–812

17. Brown, K.D. et al. (2014) Activation of SIRT3 by the NAD+ pre-cursor nicotinamide riboside protects from noise-induced hearingloss. Cell Metab. 20, 1059–1068

18. Cantó, C. et al. (2012) The NAD+ precursor nicotinamide ribosideenhances oxidative metabolism and protects against high-fat dietinduced obesity. Cell Metab. 15, 838–847

19. Barbosa, M.T. et al. (2007) The enzyme CD38 (a NAD glycohy-drolase, EC 3.2.2.5) is necessary for the development of diet-induced obesity. FASEB J. 21, 3629–3639

20. Trammell, S.A.J. et al. (2016) Nicotinamide riboside opposes type2 diabetes and neuropathy in mice. Sci. Rep. 6, 26933

21. Gariani, K. et al. (2016) Eliciting the mitochondrial unfolded proteinresponse by nicotinamide adenine dinucleotide repletion reversesfatty liver disease in mice. Hepatology 63, 1190–1204

22. French, S.W. (2016) Chronic alcohol binging injures the liver andother organs by reducing NAD+ levels required for sirtuin’s deace-tylase activity. Exp. Mol. Pathol. 100, 303–306

23. Tummala, K.S. (2014) Inhibition of de novo NAD+ synthesis byoncogenic URI causes liver tumorigenesis through DNA damage.Cancer Cell 26, 826–839

24. Guan, Y. et al. (2017) Nicotinamide mononucleotide, an NAD+

precursor, rescues age-associated susceptibility to AKI in a sirtuin1-dependent manner. J. Am. Soc. Nephrol. 28, 2337–2352

25. Tran, M.T. et al. (2016) PGC1a drives NAD biosynthesis link-ing oxidative metabolism to renal protection. Nature 531,528–532

26. Boslett, J. et al. (2017) Endothelial cells highly express CD38 withactivation by hypoxia/reoxygenation depleting NAD(P)H. Am. J.Physiol. Cell Physiol. Published online November 29, 2017. http://dx.doi.org/10.1152/ajpcell.00139.2017

27. Zhang, Y. (2016) Exogenous NAD+ administration significantlyprotects against myocardial ischemia/reperfusion injury in ratmodel. Am. J. Transl. Res. 8, 3342–3350

28. Gerdts, J. et al. (2015) SARM1 activation triggers axon degener-ation locally via NAD+ destruction. Science 348, 453–457

29. Park, J.H. et al. (2016) Nicotinamide mononucleotide inhibitspost-ischemic NAD+ degradation and dramatically amelioratesbrain damage following global cerebral ischemia. Neurobiol. Dis.95, 102–110

30. Ryu, D. et al. (2016) NAD+ repletion improves muscle function inmuscular dystrophy and counters global PARylation. Sci. Transl.Med. 8, 361ra139

31. Cerutti, R. et al. (2014) NAD+-Dependent activation of Sirt1 cor-rects the phenotype in a mouse model of mitochondrial disease.Cell Metab. 19, 1042–1049

32. Shi, H. et al. (2017) NAD deficiency, congenital malformations,and niacin supplementation. N. Engl. J. Med. 377, 544–552

33. Longo, V.D. et al. (2015) Interventions to slow aging in humans:are we ready? Aging Cell 14, 497–510

34. Aksoy, P. et al. (2006) Regulation of SIRT 1 mediated NADdependent deacetylation: a novel role for the multifunctionalenzyme CD38. Biochem. Biophys. Res. Commun. 349, 353–359

35. Aksoy, P. et al. (2006) Regulation of intracellular levels of NAD: anovel role for CD38. Biochem. Biophys. Res. Commun. 345,1386–1392

36. Chiang, S.H. et al. (2015) Genetic ablation of CD38 protectsagainst western diet-induced exercise intolerance and metabolicinflexibility. PLoS One 10, e0134927

37. Hu, Y. et al. (2014) Overexpression of CD38 decreases cellularNAD levels and alters the expression of proteins involved inenergy metabolism and antioxidant defense. J. Proteome Res.13, 786–795

38. Malavasi, S. et al. (2008) Evolution and function of the ADP ribosylcyclase/CD38 gene family in physiology and pathology. Physiol.Rev. 88, 841–886

39. Chini, E.N. (2009) CD38 as a regulator of cellular NAD: a novelpotential pharmacological target for metabolic conditions. Curr.Pharm. Des. 15, 57–63

40. Deshpande, D.A. et al. (2017) CD38 in the pathogenesis ofallergic airway disease: potential therapeutic targets. Pharmacol.Ther. 172, 116–126

41. Lokhorst, H.M. et al. (2015) Targeting CD38 with daratumumabmonotherapy in multiple myeloma. N. Engl. J. Med. 373,1207–1219

42. van de Donk, N. et al. (2018) CD38 antibodies in multiple mye-loma: back to the future. Blood 131, 13–29

43. de Weers, M. et al. (2011) Daratumumab, a novel therapeutichuman CD38 monoclonal antibody, induces killing of multiplemyeloma and other hematological tumors. J. Immunol. 186,1840–1848

44. Horenstein, A.L. et al. (2015) NAD+-metabolizing ectoenzymes inremodeling tumor–host interactions: the human myeloma model.Cells 4, 520–537

45. Krejcik, J. et al. (2017) Monocytes and granulocytes reduce CD38expression levels on myeloma cells in patients treated with dar-atumumab. Clin. Cancer Res. 23, 7498–7511

46. Feng, X. et al. (2017) Targeting CD38 suppresses induction andfunction of T regulatory cells to mitigate immunosuppression inmultiple myeloma. Clin. Cancer Res. 23, 4290–4300

Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4 435

47. Nijhof, I.S. et al. (2016) CD38 expression and complement inhib-itors affect response and resistance to daratumumab therapy inmyeloma. Blood 128, 959–970

48. Deckert, J. et al. (2014) SAR650984, a novel humanized CD38-targeting antibody, demonstrates potent antitumor activity inmodels of multiple myeloma and other CD38+ hematologic malig-nancies. Clin. Cancer Res. 20, 4574–4583

49. Martin, T. et al. (2017) A phase 1b study of isatuximab pluslenalidomide and dexamethasone for relapsed/refractory multiplemyeloma. Blood 129, 3294–3303

50. Chatterjee, S. et al. (2018) CD38–NAD+ axis regulates immuno-therapeutic anti-tumor T cell response. Cell Metab. 27, 85–100

51. Lee, H.C. and Aarhus, R. (1991) ADP-ribosyl cyclase: an enzymethat cyclizes NAD+ into a calcium-mobilizing metabolite. Cell.Regul. 3, 203–209

52. Matalonga, J. et al. (2017) The nuclear receptor LXR limits bac-terial infection of host macrophages through a mechanism thatimpacts cellular NAD metabolism. Cell Rep. 18, 1241–1255

53. Franceschi, C. and Campisi, J. (2014) Chronic inflammation(inflammaging) and its potential contribution to age-associateddiseases. J. Gerontol. A Biol. Sci. Med. Sci. 69, S4–S9

54. Shrimp, J.H. et al. (2014) Revealing CD38 cellular localizationusing a cell permeable, mechanism-based fluorescent small-molecule probe. J. Am. Chem. Soc. 136, 5656–5663

55. Grozio, A. (2013) CD73 protein as a source of extracellular pre-cursors for sustained NAD+ biosynthesis in FK866-treated tumorcells J. Biol. Chem. 288, 25938–25949

56. Liu, J. et al. (2017) Cytosolic interaction of type III human CD38with CIB1 modulates cellular cyclic ADP-ribose levels. Proc. Natl.Acad. Sci. 114, 8283–8288

57. Zielinska, W. et al. (2004) Metabolism of cyclic ADPribose: zinc isan endogenous modulator of the cyclase/NAD glycohydrolaseratio of a CD38-like enzyme from human seminal fluid. Life Sci.74, 1781–1790

58. Sauve, A.A. (2000) A covalent intermediate in CD38 is responsiblefor ADP-ribosylation and cyclization reactions. J. Am. Chem. Soc.122, 7855–7859

59. Zhang, S. et al. (2015) Comparative analysis of pharmacophorefeatures and quantitative structure–activity relationships for CD38covalent and non-covalent inhibitors. Chem. Biol. Drug Des. 86,1411–1424

60. Liu, Q. et al. (2009) Structural basis for enzymatic evolution from adedicated ADP-ribosyl cyclase to a multifunctional NAD hydro-lase. J. Biol. Chem. 284, 27637–27645

61. Horenstein, A.L. et al. (2016) Adenosine generated in the bonemarrow niche through a CD38-mediated pathway correlates withprogression of human myeloma. Mol. Med. 22, 694–704

62. Krauset, D. et al. (2014) Nicotinamide N-methyltransferaseknockdown protects against diet-induced obesity. Nature 508,258–262

63. Avalos, J.L. et al. (2005) Mechanism of sirtuin inhibition by nico-tinamide: altering the NAD+ cosubstrate specificity of a Sir2enzyme. Mol. Cell 17, 855–868

64. Smithson, G. et al. (2017) CD38+ cell depletion with TAK-079reduces arthritis in a cynomolgus collagen-induced arthritis (CIA)model. J. Immunol. 198 (Suppl. 1), 127.17

65. Liu, Z. et al. (2013) Studies on CD38 inhibitors and their applica-tion to cADPR-nediated Ca2+ signaling. Messenger 2, 19–32

66. Muller-Steffner, H.M. et al. (1992) Slow-binding inhibition of NAD+

glycohydrolase by arabino analogues of b-NAD+. J. Biol. Chem.267, 9606–9611

67. Slama, J.T. and Simmons, A.M. (1988) Carbanicotinamide ade-nine dinucleotide: synthesis and enzymological properties of acarbocyclic analogue of oxidized nicotinamide adenine dinucleo-tide. Biochemistry 27, 183–193

68. Wall, K.A. et al. (1998) Inhibition of the intrinsic NAD+ glycohy-drolase activity of CD38 by carbocyclic NAD analogues. Bio-chem. J. 335, 631–636

69. Sauve, A.A. and Schramm, V.L. (2002) Mechanism-based inhib-itors of CD38: a mammalian cyclic ADP-ribose synthetase.Biochemistry 41, 8455–8463

70. Kwong, A.K.Y. et al. (2012) Catalysis-based Inhibitors of thecalcium signaling function of CD38. Biochemistry 51, 555–564

71. Kellenberger, E. et al. (2011) Flavonoids as inhibitors of humanCD38. Bioorg. Med. Chem. Lett. 21, 3939–3942

72. Escande, C. et al. (2013) Flavonoid apigenin is an inhibitor of theNAD+ase CD38: implications for cellular NAD+ metabolism, pro-tein acetylation, and treatment of metabolic syndrome. Diabetes62, 1084–1093

73. Boslett, J. et al. (2017) Luteolinidin protects the postischemicheart through CD38 inhibition with preservation of NAD(P)(H). J.Pharmacol. Exp. Ther. 361, 99–108

74. Shu, B. et al. (2018) Blockade of CD38 diminishes lipopolysac-charide-induced macrophage classical activation and acute kid-ney injury involving NF-kB signaling suppression. Cell. Signal. 42,249–258

75. Blacher, E. et al. (2015) Inhibition of glioma progression by a newlydiscovered CD38 inhibitor. Int. J. Cancer 136, 1422–1433

76. Blacher, E. et al. (2015) Targeting CD38 in the tumor microenvi-ronment; a novel approach to treat glioma. Cancer Cell Micro-environ. 2, e486

77. Schiavoni, I. et al. (2017) CD38 modulates respiratory syncytialvirus-driven pro-inflammatory processes in human monocyte-derived dendritic cells. Immunology Published online November27, 2017. http://dx.doi.org/10.1111/imm.12873

78. Haffner, C.D. et al. (2015) Discovery, synthesis, and biologicalevaluation of thiazoloquin(az)olin(on)es as potent CD38 inhibitors.J. Med. Chem. 58, 3548–3571

79. Becherer, J.D. et al. (2015) Discovery of 4-amino-8-quinolinecarboxamides as novel, submicromolar inhibitors of NAD-hydro-lyzing enzyme CD38. J. Med. Chem. 58, 7021–7056

80. Cambronne, X.A. et al. (2016) Biosensor reveals multiple sourcesfor mitochondrial NAD+. Science 352, 1474–1477

81. Zhao, Y. et al. (2015) SoNar, a highly responsive NAD+/NADHsensor, allows high-throughput metabolic screening of anti-tumoragents. Cell Metab. 21, 777–789

82. Partida-Sanchez, D.A. et al. (2001) Cyclic ADP-ribose productionby CD38 regulates intracellular calcium release, extracellular cal-cium influx and chemotaxis in neutrophils and is required forbacterial clearance in vivo. Nat. Med. 7, 1209–1216

83. Lopatina, O. et al. (2012) The roles of oxytocin and CD38 in socialor parental behaviors. Front. Neurosci. 6, 182

436 Trends in Pharmacological Sciences, April 2018, Vol. 39, No. 4